Essays on TRIPs, pharmaceuticals and developing countries with special reference to india

While developed countries produce new drugs, developing countries usually produce generic versions of brandname drugs not only for domestic use but also for export to the rest of the wor

PART I

CHAPTER 1

GLOBAL PHARMACEUTICAL MARKETS:

BRANDNAME VS GENERIC DRUGS

In Part I of the thesis, we analyze certain aspects of the behaviour of international pharmaceutical markets The global pharmaceutical industry is undergoing rapid changes, both from the demand and supply sides The demand side dynamics reflect the affordability and access of drugs while the supply side changes correspond to innovations of new drugs through R&D While policymakers are concerned about rising costs of health care, pharmaceutical companies are seeking ways to discover new domestic and international markets to cover the accelerating costs of R&D

Estimating the demand for pharmaceutical products is challenging since demand is usually determined by the physician’s prescription for pharmaceutical products and patients are usually covered by some sort of insurance in developed countries Thus, there may be a discrepancy between the realized marginal benefits and the price paid for pharmaceutical products in developed countries In developing countries, most of the expenditure on prescription drugs is made by patients’ own funds On the supply side, the cost of producing and marketing the first unit of the drug is very high, but the marginal cost of manufacturing additional units of the drug is small

While developed countries produce new drugs, developing countries usually produce generic versions of brandname drugs not only for domestic use but also for export to the rest of the world For instance, the destination of exports

of Indian pharmaceuticals is North America, Western Europe, Japan and Australia

In 2002-03, India exported nearly 40 percent of its total pharmaceutical output with exports valued at nearly $2.6 billion The major market is the U.S which accounted for 17 percent of all Indian pharmaceutical exports in 2002-03 Moreover, the generic markets in the U.S and E.U are likely to expand at 10-15 percent per year, as attempts are made to contain rising health care costs by switching to less expensive drugs

1.1 BACKGROUND From a worldwide perspective, the pharmaceutical industry consists of a large number of firms producing a variety of drugs including brandname and generic drugs It was after World War II that the modern pharmaceutical industry emerged in its present form with its characteristic features of high research intensity and high sales promotion Large pharmaceutical multinational corporations (MNCs) co-exist with small firms in this industry The competition

in pharmaceutical markets is based on product differentiation based on R&D efforts and hence the need for patents to reward the innovator for undertaking risky ventures and spending huge resources for inventing new drugs

The reason why patents are crucial for pharmaceutical firms is that it takes hundreds of millions of dollars to discover, develop and get regulatory approval

for a new medicine In the absence of patent protection, imitators can duplicate the innovator’s chemical compound at a fraction of the latter’s cost since imitation costs are low compared to the costs of discovering and developing a new drug Moreover, with or without patent protection, the pricing policy for the product is likely to be different at each stage of the product’s life cycle Throughout the product cycle, continual changes occur in price elasticity and in costs of production and distribution, so that these changes call for price adjustments

1.1.A Patent system

The pharmaceutical industry is one of the most research-intensive manufacturing industries with product quality subjected to close regulation by regulatory authorities like the USFDA (U.S Food and Drug Administration), which regulates the introduction of new drugs and maintains high product quality standards, and approval requires difficult and lengthy pre-clinical and clinical trials Given the uncertainty in predicting the outcomes of pharmaceutical research, there is intense debate on the issue of pharmaceutical pricing, financing

of research and optimal length of patent protection Given that knowledge (including knowledge about pharmaceutical technology) is a global public good, national self-interests are often in conflict with the worldwide harmonization of patent legislations propagated by the Agreement on Trade Related Aspects of Intellectual Property Rights (TRIPs)

According to TRIPs, all member-countries of the World Trade Organization (WTO) have to provide exclusive marketing rights to an inventor

through the grant of patents for 20 years from the date of filing the patent application Patents can be either product patents or process patents Product patents are granted for a new product while process patents are granted for a novel method of manufacturing a similar or identical product In principle, the patent system provides incentives to innovators to engage huge resources in drug innovation and development and earn monopoly profits till the patent expires However, the effective patent life is 12-15 years because of the time taken in clinical testing and obtaining regulatory approval When the patent expires, generic competitors can enter the market with barriers to entry being reduced

Pharmaceutical R&D expenditures have increased sharply over the years

It has been estimated that at 1987 prices, the out-of-pocket cost for every new

chemical entity (NCE) approved was $231 million (DiMasi et al., 1991) More

recent estimates show that since then, costs have increased at an annual rate of 7.4

percent above general inflation DiMasi et al (2003) surveyed costs of drug

development of 10 large pharmaceutical firms producing 68 drugs over the period 1980-99 At 2000 dollars, they reported the average estimated base case cost of every new drug development to be $403 million and the total pre-approval capitalized cost up to the marketing stage to be $802 million Once a drug is developed and ready to be marketed, these R&D costs become sunk costs and thus become irrelevant for short-run pricing decisions The decision to carry on with such high-risk projects thus depends on potential therapeutic benefits and future revenues which are guaranteed by patents till the patent term expires

As noted above, under patent protection, brandname drugs enjoy market exclusivity for a limited time period The nominal patent life is the time between patent registry and expiration whereas the effective patent life is the time between product entry and expiration Grabowski and Vernon (2000) found that for NCEs introduced between 1990 and 1995 in the U.S., the effective patent life was only 11.7 years, despite the 20 year patent term stipulated by TRIPs

The extent to which market exclusivity for brandname drugs guarantees monopoly rights depends on the competition they face from other brands within a therapeutic class An improved competitor “me-too” drug treating the same ailment can be awarded patent rights as long as it is sufficiently different from the original drug, raising the issue of patent breadth or scope Thus, the analysis of patent protection requires a two-dimensional approach that includes patent length and patent breadth

1.1.B Process of drug innovation

New drug development is a lengthy sequential process fraught with risk and uncertainty Typically, even though the development of NCEs may be spread over many years, only a small proportion of these new products finally get approval for marketing The process of drug innovation begins with research by chemists and biologists to develop concepts for new compounds, which are then synthesized and tested for pharmacological activity If found promising, the innovator firm files for an investigational new drug application with the regulatory authority and

on approval, it goes on to the next stage of clinical trials for testing the efficacy and toxicity of the drug

DiMasi et al (2003) describe the three phases of clinical trials In phase I,

usually tests are carried out on a small number of healthy volunteers to determine the safety, absorption, metabolism and toxicity of the compound In phase II, the testing is done on more patients having the targeted condition or disease In the final phase III of clinical trials, testing is done on a large scale in different locations to establish efficacy and look for side-effects These clinical trials require testing on humans and are very time-consuming Once the phase of clinical development is successful, the innovator firm can apply for a new drug application (NDA) In general, the USFDA takes one to four years to grant approvals after reviewing the applications, but if the firm concerned is a research-intensive firm reputed for its quality, the review period may be considerably shortened (Olson, 1997) With the USFDA clearing the drug, the innovator firm can start marketing the product

1.2.A Prescription drug insurance

One of the reasons attributed for habit persistence in drug prescriptions is the insurance coverage provided for drug expenditure, particularly in developed countries In a study of the U.S., Berndt (2002) found that costs for prescription drugs increased from 5.1 percent of total health care costs in 1980 to 9.7 percent

in 2000, which translates into $121.8 billion or $450 per person At the same time,

he found that pharmaceutical expenditures grew at about 11.9 percent between

1987 to1994 and 12.9 percent between 1994 and 2000 In the first sub-period, the price growth of 6.1 percent annually was responsible for more than half the sales growth but in the second sub-period, price growth was responsible for only a fifth

of revenue growth, four-fifths being accounted by volume changes Thus, in recent years, price increases have become less important for explaining rising pharmaceutical expenditures, and instead, quantity growth in terms of greater use

of brandname products, has become the prime mover for increased spending

Nearly 40 percent of prescriptions in the U.S are covered by insurance (Scott Morton, 2000) Given that a patient is only concerned about quality and is not paying out of her own pocket for prescription drugs, it is likely that she will not purchase the cheaper generics available in the market after patent expiration Many empirical studies have found that expenditure on prescription drugs is

higher for patients with higher insurance coverage (Leibowitz et al 1985; RAND

Health Insurance Experiment, 1993; Lundin, 2000; Pavcnic, 2002; Cleanthouse,

2003 and Buchmueller et al., 2004) In fact, while insurance leads to greater

spending, growth in insurance coverage is itself stimulated by the availability of better and newer drugs (Danzon and Pauly, 2001 and 2002)

1.2.B Detailing

Sales promotion for pharmaceuticals takes several forms, the most important of which is detailing This is a costly process that requires visits by medical representatives to physicians, pharmacists and other professional distributors to promote new drugs and answer queries The large force of detailers is a fixed cost for brandname firms and acts as an incentive for continuous innovations While detailing no doubt helps in information dissemination, it also encourages brand consciousness among physicians since detailing is not aimed at the consumer but

is primarily aimed at increasing the likelihood that physicians prescribe brandname pharmaceutical products Moreover, despite large expenditures on marketing and promotion, about 20 to 30 percent of sales, promotional information seldom reports actual prices (Brekke and Kuhn, 2003) The fact that physicians do not know the price of the drugs they prescribe suggests that they cannot be fully price sensitive and at best can only estimate the magnitude of cost saving from generics Medical journal advertising is also an important promotional activity aimed at maintaining the general goodwill of the brandname firm Expenditures on detailing are four to five times higher than for journal

advertising in the U.S pharmaceutical industry (Berndt et al., 1999)

1.2.C Consumption Externalities

Besides physician prescription and advertising, consumption externalities also

play a role in affecting demand for pharmaceuticals (Berndt et al., 2003a)

Consumption externalities arise when demand for pharmaceuticals by patients and physicians depends on the number of other patients that have taken the drug and demand is influenced by their opinion regarding its efficacy, safety and acceptability Thus, herd behaviour can also create captive markets for a particular drug despite the availability of close substitutes

1.3 RISING HEALTH CARE COSTS Among all the components of health care expenditures, spending on prescription drugs is growing the fastest In the U.S in 2001, spending on prescription drugs alone exceeded $150 billion (Dubois, 2003) The primary reasons propelling this growth in health care spending are increased third-party prescription drug coverage, the introduction of newer and better products, increased outlays for

sales promotion through direct to consumer (DTC) marketing and demographic changes

1.3.A Cross-price sensitivity

Recent developments in the health sector have increased the cross-price

sensitivity between brandname and generic drugs even in the U.S Ellison et al

(1997) found a high elasticity of substitution between brandname and generic drugs Prescription drug markets in the U.S have changed considerably in recent years, due to governmental scrutiny of the drug industry and the rapid growth of managed care The health care reform initiatives in 1993 included limits on drug prices and Medicaid reimbursement, as well as other regulatory controls Also, the emergence of managed care as a major player in the prescription marketplace has recently altered the way drugs are prescribed In order to contain escalating prescription drug costs, managed care providers employ various methods to influence physicians’ prescription decisions such as drug budgets and formularies (regulated lists of approved medications) to restrict reimbursement to a fairly small number of cost-effective drugs in a given therapeutic class

Private third-party insurers have also introduced a number of measures to control the spiraling health care expenditures First, they now insist on using drugs included in their list of formularies Second, they have adopted a “two-tier” method for lower co-payments from users of generic drugs as compared to users

of brandname drugs Thus, private insurers were able to make the demand for pharmaceuticals more elastic by the threat of excluding brandname drugs from

their formulary list Also, mail order dispensing of medications to treat long-term chronic conditions is being encouraged by insurers, which in turn has exerted a pressure on retail pharmacy to lower their dispensing costs

1.3.B New drugs

Spending in a therapeutic class increases with the number of new products entering the class and the extent of their therapeutic benefit New products play a significant role in increasing drug expenditures since new products are priced at a higher level (Danzon and Ketcham, 2004) However, newer and better quality drugs tend to lower other nondrug expenditures on health care, resulting in a reduction in overall cost of treatment for a given condition (Lichtenberg, 2001)

Mullins et al (2001) studied the increase in drug spending between 1995

and 1998 and found that overall spending increased by 15 percent for all drugs, 25 percent for newer drugs and 7 percent for older drugs Thus new drugs contributed 75.3 percent of the increase in drug expenditures; the remaining 25 percent was due to expenditure growth of drugs more than five years old in the market Thus, considering expenditures on new drugs is important for understanding and predicting future drug trends

1.4 PATENT EXPIRATION AND GENERIC ENTRY Griliches and Cockburn (1996) have listed the requirements for generics to be certified as “therapeutically equivalent” by the USFDA: (a) they should contain the same active ingredients, be of the same dosage form and strength and offer the

identical route of administration while meeting the same quality standards, (b) they should be bioequivalent to the original brandname drug with respect to the rate and extent of absorption of the active ingredient and its availability at the site

of action, (c) they should be properly labeled and manufactured in compliance with good manufacturing practices However, even therapeutically equivalent products may differ in other characteristics like shape, color, flavor, packaging, labeling and shelf life These trivial factors are crucial in affecting the patient’s ability to distinguish between different pills and their dosages

It is believed that worldwide, the profitability of pharmaceutical majors is likely to be squeezed due to shrinking sales growth after patent expiration and mounting costs for R&D and advertising The total generic business is expected to grow due to expiration of patents or exclusivity for brandname drugs According

to Graham Lewis, Vice-President, Strategic Consulting, IMS Health, out of the 35 leading molecules worldwide, 13 molecules (Simvastatin, Omeprazole, Amlodipine, Fluoxetin, Loratadine, Pravastatin, Fluticasone, Amoxyclavulanic Acid, Sertraline, Lisinopril, Nifedipine, Ciprofloxacin, Famotidine) will lose patents by 2005 throwing open sales worth $52.9 billion for top 29

pharmaceutical companies (Lewis, 2001)

In the U.S., the Drug Price Competition and Patent Term Restoration Act (commonly known as the Hatch-Waxman Act) of 1984 eliminated the strict requirements for USFDA approval of generic substitutes by easing the stringent testing requirements With the passage of the Hatch-Waxman Act, the market share of generic drugs has more than doubled in the U.S from 18.6 percent in

1984 to 47.1 percent in 1999 (Berndt, 2001) More recently, the Greater Access to Affordable Pharmaceuticals Act (GAAP) in 2003 was passed in the U.S Senate to cut health care costs and close loopholes in the 1984 Act that allowed brandname drug manufacturers to keep the generic manufacturers out of the market The Congressional Budget Office has estimated that this Act would result in savings worth $60 billion over the next 10 years

1 4.A Entry procedure

In the U.S., generics producers enter the market by applying to the USFDA for an abbreviated new drug application (ANDA) They can file an ANDA even before a patent expires and so with USFDA approval can enter the market on the very day

of patent loss They have to prove that their product is exactly the same as the brandname drug in strength, dosage form and route i.e it is a “bioequivalent” But even bioequivalency studies for ANDAs can take months to years depending on the drug formulation and the generic firm’s research capacity The application process requires factory inspections and independent tests of several preliminary batches of the product Thus, the generic firm has to be ready to make the drug months before it gets the approval for marketing These entry costs are sunk costs that must be incurred prior to entry and these entry costs can be significant when there is likely to be fierce competition in the generic market

Empirical studies on the impact of generic entry on brandname drugs show that quantity demanded of brands is adversely affected with the availability

of cheaper therapeutic equivalents According to Caves et al (1991), the demand

for brandname drugs falls by 20 percent in the year of patent expiration and

continues to decline at the rate of 12 percent per year Aronsson et al (2001)

analyzed the impact of generic competition on the market shares of 12 brandname drugs from 1972 to 1996 In five out of twelve cases, entry by cheaper generics

adversely affected the market share of the brandname drug Berndt et al (2003b)

found consumers benefited from generic entry and the introduction of counter (OTC) versions of prescription drugs on patent expiration for antiulcer and heartburn drugs

over-the-Generic substitutes appear in the market at 30-50 percent lower rates and capture much of the market share of brandname drugs (Griliches and Cockburn, 1996) However, even with such large price differentials between brands and generics, the former is able to maintain a substantial share of its demand since not all patients switch to the cheaper substitute Thus, patients have different valuations or differential information regarding the quality and efficacy of such drugs resulting in considerable patient heterogeneity and physician preferential prescription of drugs

1.4.B Entry determinants

Characteristics of a drug market are important determinants of generic entry Some of the factors that attract generic entrants are larger sales revenues of the brandname firm prior to patent expiration, profit opportunities in a particular market, whether a drug treats chronic conditions as opposed to acute illness and the extent to which the drug is purchased by hospitals

Scott Morton (1999) found that the estimated costs of filing an ANDA range from $250,000 to $20 million depending on different drugs and drug-firm combinations and that generics enter brandname markets with annual sales of $22 million where the first entrant receives revenues worth less than $10 million The first entrants are the important players since late entrants often withdraw applications in response to rivals’ approvals On average, five generics enter the market in the first year of patent loss but only three more enter in the next three years (Frank and Salkever, 1997) A study by Hudson (2000) regarding generic entry across four countries, the U.K., U.S., Germany and Japan also showed that the market size at patent expiration is an important consideration, both for generic entry and the lag between patent expiration and generic entry He concludes that the patent system acts as a “risk-smoothing” device since successful brands attract greater generic entry than the less successful brands whose revenue stream will be unaffected by patent expiration

1.4.C Brand price response

The price response of brandname drugs to generic entry has been a source of controversy However, most studies show that prices of brandname drugs rise

after generic entry (Grabowski and Vernon, 1990; Caves et al., 1991; Frank and Salkever, 1992 and 1997; Suh et al., 2000; Ellison and Ellison, 2000; Danzon and

Chao, 2000) Grabowski and Vernon (1990) studied the effect of generic entry on

18 drugs between 1984 and 1987 and found that brand prices rose with generic entry They found that the brand price was higher by 7 percent one year after

generic entry and by 11 percent two years after generic entry Caves et al (1991)

also estimated a model of the ratio of generic to brandname prices for 30 drugs that went off patent during 1976-1987 and found similar results that greater generic entry, measured by the number of ANDA applications, had a downward effect on generic prices relative to brandname prices The estimated results showed that initial generic entry resulted in brand prices falling about 2 percent Frank and Salkever (1997) used data for a sample of 32 drugs during the 1980s to show that brand prices increase after generic entry In addition, they find that after three years of entry, generic prices were less than half that of brand prices This implies that the generic market is a highly competitive fringe to the brandname drug However, they concede that with generics capturing a large share of the market at significantly reduced prices, the average price of a drug subjected to generic competition falls

Suh et al (2000) found that on patent loss, innovator firms continued to

increase prices while price of generic drugs fell significantly over time By the fourth year after patent expiration, innovators’ sales had decreased 12 percent and quantity had decreased 30 percent Thus, innovators have the first-mover advantage with a captive market of brand-loyal consumers Another study by Ellison and Ellison (2000) shows that prices of brandname drugs increased (relative to a pharmaceutical price index) during the final years of patent protection After patent expiration, prices fall for the group of pharmaceuticals where entry is predicted to be likely, and continue to rise for pharmaceuticals where the probability of entry is low or medium high

Danzon and Chao (2000) found that competition from generics had little effect on the price of brandname drugs, but this may not reflect the true extent of competition because rebates are given to managed care customers to get included

in the list of formularies They found a monthly price fall of 3.7 percent for every month of entry lag indicating that price undercuts by successive entrants does not fully erode the first-mover advantage of the brandname firm They conclude that generic competition plays a greater role in lowering prices in the U.S and other less regulated markets like Canada, U.K., Germany than in the more strictly regulated markets in France, Italy and Japan

Perloff et al (2005) studied the anti-ulcer drug market and found that the

price rose with the entry of other firms They demonstrated that the effect of entry

on price depends on how close together the products are located in characteristic space They showed that if the new product is a perfect substitute to the existing product, price would fall, but if it is located far apart, then price of the existing product will not be affected If the products are located near enough to compete for the same consumers, then the existing firm will raise prices and sell to

consumers whose demand is relatively inelastic Ellison et al (1997) examined

the price sensitivity and generic entry for four cephalosporins They found high cross price elasticity among the different generics but low elasticity for brandname drugs used to treat the same conditions

To summarize, the goodwill developed by the innovator firm during the period of patent protection goes a long way in differentiating its product from later generic entrants This can be seen from the lukewarm response by brands to

generic prices and the inability of generics to swamp the market even with such large price discounts In general, one observes falling or stable trends in the prices

of generics and the simultaneous rising trends in their quantities However, given the emphasis on cost containment and the growth of managed care, the price differential between brandname and generic drugs may become lower As the cross-price sensitive segment of the market grows, the increases in brandname prices with generic entry will be dampened The expansion of the cross-price elastic segment of the pharmaceutical market would tend to reduce prices if this growth does not cause an increase in demand of brandname drugs and if the demand curve for cross-price sensitive consumers is more elastic than the demand curve for loyal consumers (Frank & Salkever, 1992)

1.5 REGIONAL DISPARITY IN HEALTH EXPENDITURES Given the disparities in health spending across different regions of the world, in general, and low purchasing power of consumers in developing countries, in particular, there continues a raging debate on the extension of pharmaceutical product patent rights to developing countries Table 1.1 presents the year of adoption of product patents for pharmaceuticals for some developed and developing countries with their GDP per capita at that time It is clear that pharmaceutical patent protection was adopted in the developed world recently after reaching a high level of per capita income However, developing countries now extending protection to pharmaceutical products are at a much lower level of income than those adopting earlier This adversely affects the welfare of poor

consumers in low-income countries arising from higher prices due to the adoption

of stricter patent laws In fact, the Doha Declaration on TRIPs and Public Health has extended the deadline for the implementation of pharmaceutical patent protection in 49 least developed countries from 2006 to 2016 These recent steps implicitly recognize that the trade-offs between pricing and incentives associated with patents are different for different countries

Undoubtedly patents stimulate R&D investments for lifesaving drugs, but

at the same time patents also raise pharmaceutical prices for consumers in income countries, denying them access to medicine The incentives to invest in research on global diseases in developed and developing countries are different from those for neglected diseases (like malaria) that primarily affect developing countries Witness the case that the USFDA licensed only 8 out of 1,233 drugs (less than one percent) between 1975 and 1997 for tropical diseases (Lanjouw, 2003) For global diseases, the prime mover for R&D is the market in developed countries so that extending patent rights to developing countries only raises prices while contributing little to incentives Thus, the main policy issue is to stimulate R&D for “orphan drugs”1 and neglected diseases2 since purchasing power in poor countries is low with very little spending on health care

low-Table A1.1 lists the health expenditures of 191 WHO member-countries It clearly shows the wide disparities in payments for health expenditures between

1

Orphan drugs are defined as drugs that are developed for orphan diseases affecting less than 200,000 people These treat conditions that occur rarely and the pharmaceutical industry has little incentive to venture into these drugs owing to financial considerations

2

Neglected diseases are mainly tropical diseases that affect millions of people in low-income countries But efficacious treatment for these diseases is absent since poor people do not have the purchasing power to buy those drugs and again the pharmaceutical industry ignores these diseases due to small market size and low profitability

the developed and developing countries In some countries like the U.S., South Africa, France and the Netherlands, the share of private insurance in private health expenditure is above 50 percent But, in most middle- and low-income countries, there is negligible or no insurance plan at all for covering health expenditures In general, most of the health expenditure by patients in these countries is met by out-of-pocket payments, implying that the price of essential drugs does matter to poor people and to poor countries Moreover, a high-ability individual is more likely to be fully insured, while a low-ability individual only partly insured Thus, lower income patients are at a disadvantage when it comes

to insurance coverage

With low purchasing power, sales of drugs in developing countries have remained much below the level found in affluent countries In 2000, per capita expenditures on pharmaceuticals in North America, Europe and Japan were huge

at $484.2, $204.6 and $455.7, respectively as compared to the per capita sales of only $2.7 in the Indian subcontinent and $6.3 in East Asia including China (Table 1.2) Growth estimates for 2005 also reflect the same scenario with North America, Europe and Japan accounting for 83 percent of the global sales, and Southeast Asia, China, India, Latin America and the rest of the world including Africa contributing to only 17 percent of sales despite being the most densely populated regions of the world (Figure 1.1)

1.5.A Reform of the patent system

Even after the signing of TRIPs in 1994 and its enforcement in developing countries by January 2005, there is a question mark on the appropriateness and equity/efficiency effects of a uniform patent system across the world Given that developing countries and the least developed countries not only have low purchasing power but are also home to many specific diseases and epidemics, different proposals have been put forward for the reform of the patent system These include the proposal for foreign filing licenses by Lanjouw (2003) and differential pricing by Danzon (2002) In the next chapter, we analyze a proposal along these lines for exempting developing countries from provisions of TRIPs

Lanjouw (2003) proposes the foreign filing license solution such that when a pharmaceutical firm files patents for drugs related to global diseases, the patent owner cannot claim patent rights worldwide and is required to choose patent protection either in high-income countries or in low-income countries, but not both This would typically lead the patent owner to maintain patent protection

in the former rather than the latter thereby allowing access to generic drugs at lower prices in developing countries However, when the pharmaceutical firm files patents for orphan drugs related to neglected diseases, the patent owners would be provided protection worldwide to encourage R&D Thus, this proposal allows firms to protect their important markets while at the same time allowing patent rights in developing countries to provide incentives for innovation

1.5.B Differential pricing

Another policy measure advocated for access to affordable drugs in developing countries is differential pricing for pharmaceuticals (Danzon, 2002) Differential pricing entails charging different prices in different markets based on “Ramsey pricing” such that prices are inversely related to the elasticity of demand Price differentials could be such that prices in high-income countries exceed the costs

of production and distribution to cover joint costs of R&D and prices in income countries cover smaller markups In the absence of parallel imports where drugs cannot be resold in other countries at cheaper prices and given that demands are independent for patients in high- and low-income countries, it is reasonable to charge different prices in the two markets No doubt, differential pricing and ban

low-on parallel trade3 may seem to go against the standard principles of free trade and marginal cost pricing, but, in the case of research-based pharmaceuticals, joint costs of R&D cannot be sustained under marginal cost pricing Thus, from the viewpoint of efficiency and equity, differential pricing can be welfare-improving

if more new drugs are developed and distributed in developing countries and lower prices are charged in low-income countries As opposed to this, uniform pricing leads to high prices in developing countries or launch delays or even no launches at all In fact, van Dijk (1995) in a model of patent breadth generating linear demand curves shows that dynamic efficiency improves with price discrimination compared to uniform pricing

Demand heterogeneity coupled with low marginal costs of production and marketing efforts at brand consciousness creates the right setting for nonuniform pricing (Berndt, 2002) Malueg and Schwartz (1994) demonstrate that when demand dispersion across markets is large, uniform pricing by a monopolist yields lower global welfare than third-degree price discrimination4 Further, they found that “mixed systems” permitting discrimination across but not within specific groups of markets, result in greater welfare than uniform pricing or unrestricted multimarket discrimination They conclude against allowing parallel imports and support international price discrimination that allows serving additional markets

Moreover, differential pricing has long been a feature of the pharmaceutical industry Cash-paying patients pay higher prices than institutional purchasers and “the law of one price” does not hold for the prescription drug market (Frank, 2001) In addition, with unequal income distributions within some countries, a small high-income subgroup dominates potential drug sales In such cases, one needs to examine the separability of submarkets within a country say,

by designing programmes that serve only the low-income subgroup and not the high-income subgroup In this way, prices could differ between subgroups nationally and cross-nationally Jack and Lanjouw (2003) using a public economics framework for the pricing of pharmaceuticals, show that in the face of the extremely unequal distribution of world income, even the standard Ramsey pricing for covering own marginal cost should not be applied for poor countries

4

Third-degree price dicrimination occurs when a manufacturer is able to segment its customers into two or more separate markets, each market defined by its unique characteristics and price elasticity

Our proposal outlined in the next chapter is the scenario where developing countries are exempt from TRIPs obligations to enforce product patents for pharmaceuticals We build a model of pharmaceutical markets in the light of a patent race among competing firms The incentive for R&D is the patent on either the breakthrough or the me-too drug A feature of our model that has not been analyzed before is the prevalence of insurance in developed countries as opposed

to developing countries, such that the true burden of financing R&D falls to a greater extent on the former than the latter We suggest that generics drugs be allowed in low-income countries, particularly since most of them do not have a well-established and functioning pharmaceutical industry Our model makes a contribution analogous to Lanjouw (2003) and examines how valuable the implementation of this proposal would be by undertaking some simulation exercises

Table 1.1: GDP level on adoption of pharmaceutical product patents by

developed and developing countries

Country Year of Adoption GDP per capita

(1995 $)

OECD Countries

Japan 1976 24,043 Switzerland 1977 36,965 Italy 1978 13,465 Holland 1978 20,881 Sweden 1978 21,896 Canada 1983 16,296 Denmark 1983 28,010 Austria 1987 25,099 Spain 1992 14,430 Portugal 1992 10,469 Greece 1992 10,897 Norway 1992 30,389

Developing Countries

China 1992 424 Brazil 1996 4,482 Argentina 2000 8,174 Uruguay 2001 6,193 Guatemala 2000 1,563 Egypta 2005 1,250 Pakistan 2001 517

Source: Lanjouw (2003) and updated by author based on newspaper reports and

World Development Indicators CD-ROM, 2004

Notes:athe value of GDP per capita is for the year 2002

Table 1.2: Region-wise per capita pharmaceutical expenditure in 2000

Region Sales

($ billion)

Population (thousands)

Per capita sales ($) North America 152.04 313987 484.22

Europe 79.64 389282 204.58 Eastern and Central Europe 7.24 338022 21.42

Japan 57.92 127096 455.72 East Asia and China 18.1 2884708 6.27

Indian subcontinent 3.62 1329890 2.72

Latin America 25.34 518934 48.83

Rest of the world 18.1 154796 116.92

Source: Author’s own calculations based on Lewis (2001) and UNCTAD

Handbook of Statistics, 2002

Figure 1.1: Region-wise distribution of world pharmaceutical market valued

at $ 561 billion in 2005

North America 49%

Europe 22%

Latin America 7%

Japan 12%

Southeast Asia and China 5%

Rest of World 4%

Indian Subcontinent 1%

Source: Lewis (2001)

APPENDIX A1

Table A1.1: National health account estimates of 191 WHO member-states,

1997, revised as on May 31, 2001 (percentage shares)

THE/GDP PHE/THE PvtHE/THE PvtIns/PvtHE OOPS/PvtHE

United States of America 13 45.5 54.6 60.6 28.2

Source: World Health Organization (2002), Mobilization of Domestic Resources

for Health, Report of Working Group 3 of the Commission on Macroeconomics

and Health, WHO: Geneva

Notes: HE=Health Expenditure, T=Total, P=Public, Pvt=Private, Ins=Insurance,

OOPO=Out of Pocket Spending

While knowledge is a public good, patent legislations for rewarding innovations are national In order to do away with the national incentive to free-ride on R&D conducted elsewhere in the world, TRIPs was signed in 1994 under the aegis of the WTO Before 1995, many developing economies had different patent regimes, often involving weaker forms of patent legislation and shorter periods of protection, than in the developed countries However, TRIPs regulations call for a harmonization of patent laws such that developing countries

are required to enforce the same patent laws as prevalent in developed countries which include both product and process5 patents together with a longer patent term

We take the stylized facts regarding the current system to be as follows Most of the R&D that leads to a flow of new drugs is undertaken by a small group

of large multinational firms, owned principally by investors in developed countries The firms compete with each other, and when one of them achieves a

“breakthrough” leading to the discovery of a significant new drug, the others try

to produce their own versions or “me-too” drugs in the same therapeutic class However, the me-too drugs are sufficiently different from the breakthrough drugs

so that they can be patented in their own right6

In the model, we assume that in the race for patents, the assumption of

‘winner-takes-all’ does not apply While it is true that for some drugs, the R&D process will involve a leader and follower, in our model we focus on a race between two similar firms where the winner is determined by chance and the loser exploits the accumulated knowledge to engage in further research to develop a similar (patentable) drug

Patent legislation in the developed countries ensure that the firms owning the patents have monopoly rights for a period of 20 years, although for breakthrough drugs, the period of effective patent protection is smaller as there is

This brings into focus the breadth of the patent which is defined as the area in a vertically

differentiated product space that is protected by a patent Thus, it is in the interest of the innovator

to claim the broadest scope of patent protection However, if some aspects of the me-too drug resemble the breakthrough drug too closely, it opens the possibility of for the latter to initiate patent infringement cases against the latter In our model, we disregard this possibility

a substantial lag between the time when the relevant patent is registered and the time when the drug can be marketed7 When the product patent on the breakthrough drug expires, other firms are free to use the technology to produce a generic version that is protected by process patents Typically, because generic versions of brandname drugs must be aggressively marketed, the market for the generic version tends to be dominated by a single seller for a period of time after the product patent has expired8 Finally, patent legislation in some countries like the U.S and some economic regions like the E.U prohibits parallel imports (while parallel imports are allowed among member-states of the E.U., the E.U treaty does not allow parallel imports of products outside the E.U.) In our analysis, we group all developed countries together as one market (with a single price) and similarly for all developing countries Consistent with this pattern, we assume that the rules in developed countries are such that parallel imports from developing countries are ruled out With respect to parallel import rules in developing countries, these are irrelevant for our model since prices in the developed countries always are higher than in the developing world Hence the

patent owner in either market faces no effective competition from independent

imports from low-price to high-price markets by retailers

Economic efficiency in this system depends in a complex way on the length of patent protection, the breadth or scope of patents that determine how different a new version of a drug has to be in order to be patented as a different drug, and on the nature of competition among different versions of a drug As has

been well described in the literature, economic efficiency in this market involves

a tension between static efficiency (which is promoted by pricing drugs at their marginal cost of production once they have been invented) and dynamic efficiency (ensuring that the expected profitability of inventing new drugs is high enough which requires prices above marginal production cost) The tension arises because the implicit optimum is inherently second-best in comparison with one in which marginal-cost pricing is used to attain static efficiency, and incentive to undertake R&D is provided through public subsidies Thus, the rationale for strengthening the international patent system through TRIPs lies in economic efficiency, such that by increasing the potential profitability of the worldwide marketing of new drugs, it enhances dynamic efficiency by providing incentives for pharmaceutical companies to undertake R&D

However, in this paper, we argue that this Agreement in its simple form is rather tenuous In fact, it has been shown that even though monopoly rents should induce inventive efforts, the costs of disclosure of the details of the patented

innovation can more than offset the gains from patenting (Arora et al., 2003)

Further, the effects of “stronger” patent rights imply that the patent for rivals also become stronger For instance, increasing the patent scope does not necessarily result in higher expected rents since a rival may forestall the ability of a firm to commercialize its innovation (Jaffe, 2000; Gallini, 2002) This is particularly true

of pharmaceuticals where firms are engaged in patent races to develop drugs in the same therapeutic class Moreover, it has been argued by Merges and Nelson (1990) and Scotchmer (1991) that broader patents may even decelerate the rate of

innovation by curtailing subsequent innovations when the technology is cumulative and builds on existing technology (as is the case with generic pharmaceuticals) In addition, for cumulative technologies, firms may even file patents for strategic reasons so that they are able to negotiate access to other firm’s technology on more favourable terms (Hall and Zeidonis, 2001)

In this paper, we add to the literature on patents by building a model of a patent race for patents where the world is separated into two markets for developed and developing countries If a given firm wins the race in developing a new drug, it receives product patents for that drug and the other firms are left to develop the me-too and generic versions of the drug The paper also focuses on the issue of patent scope through the entry of me-too drugs, which are newer and better quality drugs in the same therapeutic class While the incentive for investing in R&D comes from the patents granted to the new drug, we show that for developing countries, exclusive marketing rights granted by patents lead to welfare losses as compared to a world without patents In fact, we go a step further and try to put some numbers on the welfare losses due to patents through simulations Finally, we demonstrate that under reasonable assumptions, other methods to enhance the profitability of pharmaceutical R&D (such as differential pricing and public subsidies) may be more efficient Moreover, because other things being equal, the larger static efficiency losses under the TRIPs affect individuals in low-income countries, they will lead to a highly unfavourable impact on the global distribution of real income

The rest of the paper is organized as follows In section 2.2, we specify the model and solve for the equilibrium configuration on the assumption that there is

a harmonization of patent laws according to TRIPs so that patent protection is similar in both developed and developing countries In section 2.3, we find the solution under the alternative assumption where developing countries have a softer patent legislation that allows generics to compete with the breakthrough drug from the outset and show the difference this makes to the returns to pharmaceutical research, and to the consumer surplus in developed and developing countries Section 2.4 is an extension where we endogenize the time taken for a me-too drug to enter the market Section 2.5 concludes with some policy suggestions

2.2 MODEL WITH TRIPs Our analysis is based on a three-period model of the lifecycle of a breakthrough

drug, starting from the time it is first marketed Let firm A be the one to develop the breakthrough drug In period 1, firm A enjoys monopoly power, since there is

no other “similar” drug but competing firms carry on working to develop their own drugs in this therapeutic class The first period comes to an end when additional versions of the drug have been patented and begin to compete with the original one We represent these me-too versions by a single drug, developed and produced by firm B In period 2, firms A and B engage in a Bertrand

competition with differentiated products Period 3 begins when the patent on the breakthrough drug expires At this point, an early generic competitor, firm G,

enters the market that is covered by process patent Period 3 lasts till the process patent on the generic drug expires and there is free entry in the market Thus, in all three periods, the drugs are protected by some form of intellectual property rights (IPRs) because even if the product patent expires, the brandname drugs are still covered by trademarks

In each period, the different drugs are marketed throughout the world For simplicity, we distinguish only two sub-markets, one in the developed world (denoted by H), and the other in the developing world (denoted by L) In each

market, individuals buy either one or zero units Individuals’ willingness to pay (henceforth WTP) for the original breakthrough drug is denoted by X, and ranges

from zero to X H in H and from zero to X in L Again for simplicity, we assume L

that potential buyers are uniformly distributed on the intervals { H}

is λHwhile the size of the market in L is denoted by λL

A key assumption in the paper is that demand conditions are such that

L H

9

The RAND Health Insurance Experiment headed by Newhouse (1993) showed that patients with greater insurance coverage bought more prescription drugs For other empirical studies that found